1
0
mirror of https://github.com/searxng/searxng.git synced 2024-11-18 02:10:12 +01:00
searxng/searx/engines/soundcloud.py

104 lines
3.0 KiB
Python
Raw Normal View History

# SPDX-License-Identifier: AGPL-3.0-or-later
"""
Soundcloud (Music)
"""
2014-09-02 18:12:30 +02:00
import re
2013-10-17 21:21:23 +02:00
from json import loads
2016-11-30 18:43:03 +01:00
from lxml import html
from dateutil import parser
from urllib.parse import quote_plus, urlencode
from searx.network import get as http_get
2016-11-30 18:43:03 +01:00
# about
about = {
"website": 'https://soundcloud.com',
"wikidata_id": 'Q568769',
"official_api_documentation": 'https://developers.soundcloud.com/',
"use_official_api": True,
"require_api_key": False,
"results": 'JSON',
}
2013-10-17 21:21:23 +02:00
2014-09-02 18:12:30 +02:00
# engine dependent config
2013-10-17 21:21:23 +02:00
categories = ['music']
2014-09-02 18:12:30 +02:00
paging = True
2013-10-17 21:21:23 +02:00
2014-09-02 18:12:30 +02:00
# search-url
# missing attribute: user_id, app_version, app_locale
url = 'https://api-v2.soundcloud.com/'
search_url = (
url + 'search?{query}'
'&variant_ids='
'&facet=model'
'&limit=20'
'&offset={offset}'
'&linked_partitioning=1'
'&client_id={client_id}'
) # noqa
2016-11-30 18:43:03 +01:00
cid_re = re.compile(r'client_id:"([^"]*)"', re.I | re.U)
guest_client_id = ''
2016-11-30 18:43:03 +01:00
2013-10-17 21:21:23 +02:00
def get_client_id():
response = http_get("https://soundcloud.com")
if response.ok:
2016-11-30 18:43:03 +01:00
tree = html.fromstring(response.content)
# script_tags has been moved from /assets/app/ to /assets/ path. I
# found client_id in https://a-v2.sndcdn.com/assets/49-a0c01933-3.js
script_tags = tree.xpath("//script[contains(@src, '/assets/')]")
app_js_urls = [script_tag.get('src') for script_tag in script_tags if script_tag is not None]
# extracts valid app_js urls from soundcloud.com content
for app_js_url in app_js_urls[::-1]:
# gets app_js and searches for the clientid
response = http_get(app_js_url)
if response.ok:
cids = cid_re.search(response.content.decode())
if cids is not None and len(cids.groups()):
return cids.groups()[0]
logger.warning("Unable to fetch guest client_id from SoundCloud, check parser!")
return ""
def init(engine_settings=None):
global guest_client_id
# api-key
guest_client_id = get_client_id()
2014-09-02 18:12:30 +02:00
# do search-request
2013-10-17 21:21:23 +02:00
def request(query, params):
2014-01-30 01:50:15 +01:00
offset = (params['pageno'] - 1) * 20
2014-09-02 18:12:30 +02:00
params['url'] = search_url.format(query=urlencode({'q': query}), offset=offset, client_id=guest_client_id)
2014-09-02 18:12:30 +02:00
2013-10-17 21:21:23 +02:00
return params
2014-09-02 18:12:30 +02:00
# get response from search-request
2013-10-17 21:21:23 +02:00
def response(resp):
results = []
search_res = loads(resp.text)
2014-09-02 18:12:30 +02:00
# parse results
2013-10-17 21:21:23 +02:00
for result in search_res.get('collection', []):
2013-10-20 00:52:32 +02:00
if result['kind'] in ('track', 'playlist'):
uri = quote_plus(result['uri'])
res = {
'url': result['permalink_url'],
'title': result['title'],
'content': result['description'] or '',
'publishedDate': parser.parse(result['last_modified']),
'iframe_src': "https://w.soundcloud.com/player/?url=" + uri,
}
img_src = result['artwork_url'] or result['user']['avatar_url']
if img_src:
res['img_src'] = img_src
results.append(res)
2014-09-02 18:12:30 +02:00
2013-10-17 21:21:23 +02:00
return results